Re-evolution of the 2-phenylquinolines: Ligand-based design, synthesis, and biological evaluation of a potent new class of staphylococcus aureus NorA efflux pump inhibitors to combat antimicrobial resistance

Article abstract of DOI:10.1021/jm400262a

Overexpression of efflux pumps is an important mechanism by which bacteria evade the effects of antimicrobial agents that are substrates. NorA is a Staphylococcus aureus efflux pump that confers reduced susceptibility to many structurally unrelated agents, including fluoroquinolones, biocides, and dyes, resulting in a multidrug resistant (MDR) phenotype. In this work, a series of 2-phenylquinoline derivatives was designed by means of ligand-based pharmacophore modeling in an attempt to identify improved S. aureus NorA efflux pump inhibitors (EPIs). Most of the 2-phenylquinoline derivatives displayed potent EPI activity against the norA overexpressing strain SA-1199B. The antibacterial activity of ciprofloxacin, when used in combination with some of the synthesized compounds, was completely restored in SA-1199B and SA-K2378, a strain overexpressing norA from a multicopy plasmid. Compounds 3m and 3q also showed potent synergistic activity with the ethidium bromide dye in a strain overexpressing the MepA MDR efflux pump.

Full text of DOI:10.1021/jm400262a

Article
pubs.acs.org/jmc
Re-evolution of the 2‑Phenylquinolines: Ligand-Based Design,
Synthesis, and Biological Evaluation of a Potent New Class of
Staphylococcus aureus NorA Efflux Pump Inhibitors to Combat
Antimicrobial Resistance
Stefano Sabatini,*,† Francesca Gosetto,† Nunzio Iraci,† Maria Letizia Barreca,† Serena Massari,†
Luca Sancineto,† Giuseppe Manfroni,† Oriana Tabarrini,† Mirjana Dimovska,‡ Glenn W. Kaatz,‡
and Violetta Cecchetti†
†Dipartimento di Chimica e Tecnologia del Farmaco, Universita
̀
degli Studi di Perugia, 06123 Perugia, Italy
‡John D. Dingell Department of Veterans Affairs Medical Center and the Department of Internal Medicine, Division of Infectious
Diseases, School of Medicine, Wayne State University, Detroit, Michigan 48201, United States
S
* Supporting Information
ABSTRACT: Overexpression of efflux pumps is an important
mechanism by which bacteria evade the effects of antimicrobial
agents that are substrates. NorA is a Staphylococcus aureus
efflux pump that confers reduced susceptibility to many
structurally unrelated agents, including fluoroquinolones,
biocides, and dyes, resulting in a multidrug resistant (MDR)
phenotype. In this work, a series of 2-phenylquinoline
derivatives was designed by means of ligand-based pharmaco-
phore modeling in an attempt to identify improved S. aureus
NorA efflux pump inhibitors (EPIs). Most of the 2-
phenylquinoline derivatives displayed potent EPI activity against the norA overexpressing strain SA-1199B. The antibacterial
activity of ciprofloxacin, when used in combination with some of the synthesized compounds, was completely restored in SA-
1199B and SA-K2378, a strain overexpressing norA from a multicopy plasmid. Compounds 3m and 3q also showed potent
synergistic activity with the ethidium bromide dye in a strain overexpressing the MepA MDR efflux pump.
lacking risk factors such as recent exposure to the health care
system. This increase has been associated with the recognition of
new MRSA strains, often called community-associated MRSA
(CA-MRSA), that have been responsible for a large proportion of
the increased disease burden. CA-MRSA infections, far from
being the clinical curiosity they were in the mid-1990s, have
become commonplace and have created a public health crisis in
U.S. emergency departments and other clinical settings.12 MRSA
often express concomitant resistance to many antibacterials such
as tetracyclines, aminoglycosides, and fluoroquinolones. More-
over, since 2002, vancomycin resistant strains have also been
identified.13 Such strains amplify the difficulty in treating clinical
infections related to them.
Among the different mechanisms by which bacteria may
acquire resistance to antimicrobial agents, one of the most
important involves overexpression of efflux pumps.14 Efflux
systems are responsible for the extrusion of toxic compounds,
including antibacterials, encountered in the bacterial environ-
ment.15 In a recent work, the increased expression of one or more
MDR efflux pump genes was identified in 151/309 S. aureus
INTRODUCTION
■
In the past few decades, the rate of antimicrobial resistance has
increased dramatically and, in recent years, bacterial infections
have become increasingly difficult to treat.1,2 The rise of bacterial
strains that are developing resistance to more than one first line
drug, i.e., multidrug resistant (MDR) strains, is one of the major
worldwide public health emergencies.3,4
Epidemic antibiotic resistance has been described for
numerous pathogens in varying contexts, including, but not
limited to, a global pandemic of methicillin-resistant Staph-
ylococcus aureus (MRSA) infection.5,6 S. aureus is a bacterium
responsible for a wide range of infections from simple skin and
soft tissue infections to more serious illnesses like pneumonia,
endocarditis, and sepsis.7 Among the resistant strains, MRSA is
one of the most challenging to treat.8 About 19000 deaths yearly
are attributable to MRSA in the United States.9 This exceeds the
total number of deaths related to human immunodeficiency
virus/AIDS in the U.S. in 2005.10 In the EU in 2008, MRSA
accounted for 44% (n = 171200) of healthcare-associated
staphylococcal infections and 22% (n = 5400) of attributable
extra deaths.11
Since the mid-1990s, there has been an explosion in the
number of MRSA infections in the U.S. among populations
Received: February 20, 2013
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
phenyl-4H-chromen-4-one moiety, a common feature of natural
flavone and flavonolignane EPIs,35−37 to obtain the 2-phenyl-
quinolone nucleus first, and then, after suitable substitutions, the
2-(4′-propoxyphenyl)quinoline derivatives.41 In particular, the
C-4 O-ethyl-2-diethylamine and O-ethyl-2-piperidine 2-phenyl-
quinoline derivatives (compounds 141 and 2,41 respectively;
Figure 1) resulted in very potent NorA inhibitors, showing 93.4%
and 88.5% inhibition of ethidium bromide (EtBr) efflux,
respectively, when tested at a 50 μM concentration against SA-
1199B (norA+/A116E GrlA).
Taking these interesting results into account and in an effort to
obtain more potent NorA EPIs, we have decided to expand the
SAR of this class of EPIs by introducing an O-alkyl or different O-
alkylamino chains at the C-4 position of the 2-(4′-
propoxyphenyl)quinoline nucleus.
Figure 1. Re-evolution of the 2-(4′-propoxyphenyl)quinolines EPIs.
clinical strains (49%) by quantitative reverse transcription-PCR.
Among those overexpressing a single gene, norA was the most
common (43%), followed by norB (23.2%) and mepA (9.9%).16
NorA is the most studied S. aureus efflux pump. It is a MDR
pump belonging to the major facilitator superfamily (MFS)17
and is capable of extruding multiple structurally dissimilar
substrates such as hydrophilic fluoroquinolones (norfloxacin and
ciprofloxacin) and various biocides and dyes through an antiport
mechanism driven by the proton-motive force.18,19 To date, the
structural biology of NorA has not been determined. Never-
theless, the sequence homology and the sharing of several
substrates with other MDR pumps have led to the hypothesis
that NorA may have a large hydrophobic binding site. This
structural peculiarity could explain the broad substrate specificity
of MDR pumps.20
Efflux pumps are thus viable antibacterial targets and
identification and development of potent EPIs is a promising
and valid strategy.21 There are a number of reasons to pursue this
area, including: (i) this is a nature-inspired approach, as there are
examples in which plants produce both antibacterials and EPIs
that improve the antibacterial activity (e.g., berberine and the
methoxyhydnocarpin, isoflavones and α-linolenic acid, and
dalversinol A and a methoxychalcone),22−24 (ii) EPIs used in
combination with antibiotics may not only increase antibacterial
potency25 but also may expand the antibacterial spectrum and
reduce the frequency of the emergence of target-based
resistance,26 (iii) EPIs have been shown to reduce biofilm
formation and block the antibacterial tolerance of bacterial
biofilms.27
SYNTHESIS OF A FIRST SET OF QUINOLINE
DERIVATIVES
■
To confirm the role of the protonable nitrogen of the O-
alkylamino chain at the C-4 position of 141 and 2,41 we designed
and synthesized a first set of seven compounds bearing an O-alkyl
(3a) or suitable O-alkylamines (3b, 3c, 3j, 3t, 3w, and 3y)
differing in size and degree of basicity. To assess the NorA EPI
function, compounds 3a−c, 3j, 3t, 3w, and 3y were screened for
their EtBr efflux inhibition activity at a concentration of 50 μM
against SA-1199B. If the screening EPI activity was >80%,
minimum inhibitory concentrations (MICs) were determined
and a series of concentrations was tested to evaluate the 50%
inhibitory concentration (IC50). 141 and reserpine were included
as reference compounds. The results of these assays are provided
in Table 1 and Figure 2.
Table 1. EtBr Efflux Inhibition Activity (%) at 50 μM
Concentration, IC50, and MICs of the First Set of 2-
Phenylquinoline Derivatives versus SA-1199B (norA+/A116E
GrlA)
Many EPIs capable of potentiating the activity of antimicrobial
pump substrates have been identified. Early EPIs included
reserpine and verapamil, but concentrations required for pump
inhibition activity are too high for them to be clinically
relevant.28,29 Several other nonantibiotic compounds, such as
omeprazole, paroxetine, chlorpromazine, and respective deriva-
tives, and the synthetic acridone derivative GG918, have been
shown to increase the antibacterial potency of pump substrates
by inhibiting NorA.30−34 Great efforts to identify EPIs have been
made by phytochemists who have reported series of natural EPI
compounds, extremely varied in terms of their chemical class and
shape, including flavones, isoflavones, porphyrin phaeophorbide
A, and acylated glycosides.35,36 To date, there are just a few
examples of rationally designed NorA inhibitors, and very little
work has been done with respect to their structure−activity
relationships (SAR).37−46 Although the therapeutic utility of
EPIs has yet to be validated in the clinical setting, this approach
holds promise for improving the efficacy and/or extending the
clinical utility of existing antibacterials, giving new life to old
drugs with secure social and economic benefit.25
In an attempt to obtain new chemotypes endowed with good
inhibitory activity against NorA, we have recently modified the 2-
B
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
LIGAND-BASED PHARMACOPHORE MODEL FOR
NORA EPIS
■
Because of the lack of structural biology information for NorA
and its interaction with inhibitors and/or substrates, we decided
to develop a common feature three-dimensional (3D)
pharmacophore model for highly active NorA inhibitors by
means of the software Phase.47 The main goal of this study was to
determine the essential 3D structural requirements that are
relevant for interaction with NorA. The data set for the
pharmacophore development was built by considering single
or chemical classes of compounds tested as NorA inhibitors
synthesized by us or reported in the literature by other research
groups.39−44,48,49 Our EPI library included the first set of seven
quinoline derivatives described herein (i.e., 3a−c, 3j, 3t, 3w, and
3y), 3-phenyl-1,4-benzothiazine derivatives,39 the 6-amino-7-
thiopyranopyridinyl quinolone ethyl esters,40 2-(4′-
propoxyphenyl)quinoline derivatives,41 pyrazolo[4,3-c][1,2]-
benzothiazines 5,5 dioxide,42 and other compounds recently
published.43,44
Examination of the literature revealed a large number of
compounds having the biological behavior appropriate for our
purposes. However, to ensure high homogeneity in the biological
data within the data set, we selected only compounds whose
NorA inhibitory activity was evaluated using protocols
comparable to the ones used to test our compounds. In
particular, three constraints were applied in such selection: (i)
Figure 2. Dose−response EtBr efflux inhibition assays for 3b, 3y, 1,41
and reserpine against SA-1199B (norA+/A116E GrlA).
Only the compounds carrying a small protonable O-amino-
alkyl chain at the C-4 position of the quinolone nucleus (3b and
3y) possessed good NorA EPI activity. In fact, the presence of an
O-alkyl chain (3a), a less protonable (3c, 3j, and 3w) or bulky O-
alkylamino chain (3t) at the C-4 position resulted in weak or
absent NorA EPI activity.
The good EtBr efflux inhibition activities of compounds 3b
and 3y (92.0 and 96.1%, respectively) were also confirmed by
their IC50 values, which were comparable to those of the
reference compounds in a dose−response assay (Figure 2).
These data highlighted the key role of the protonable O-
alkylamino chain at the C-4 position of the 2-(4′-
propoxyphenyl)quinoline nucleus to obtain potent NorA-
mediated EtBr efflux inhibition.
COMPUTER-AIDED DESIGN OF NEW QUINOLINE
NORA EPIS
■
On the basis of these data, we elected to synthesize additional
quinoline derivatives bearing suitable protonable side chains at
the C-4 position, with the aim of expanding our SAR knowledge
and improving NorA EPI activity for this class of EPIs. For this
purpose, a variety of computational approaches were combined
into a multistep procedure (Figure 3) to rationally support the
chemical synthesis of new quinoline NorA EPIs, thus minimizing
efforts and costs.
Figure 3. Workflow for the identification of the virtual hits and new
Figure 4. Geometry of the pharmacophore: (A) distance and (B) angles
EPIs.
between pharmacophoric features of APRR hypothesis.
C
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Table 2. The Active Set of Compounds Used to Build the Pharmacophore Model and Their Corresponding EPI Activities at 50 μM
against SA-1199B (norA+/A116E GrlA)
the NorA inhibitory activity had to be evaluated at a
concentration of 50 μM and reported as the percentage of
reduction of EtBr efflux,50,51 (ii) the bacterial strain used for these
experiments had to be SA-1199B, which overexpress the NorA
efflux pump and is characterized by a mutation of topoisomerase
IV that reduces its affinity to ciprofloxacin (CPX), and (iii) the
compounds should have well-defined chirality.
On the basis of these requirements, a total of 65 compounds
constituted our data set. The pharmacophore model was
developed by means of the Phase47 software. The computational
method we used required classification of molecules according to
activity. As such, arbitrary activity ranges were established.
Molecules possessing EtBr efflux inhibition of >70% at a 50 μM
concentration were considered “active”, and those demonstrat-
ing <30% inhibition were considered “inactive”. Following these
rules, the active set included 14 compounds (Table 2), while the
remaining 51 compounds were included in the inactive set
(Table S-1 in Supporting Information).
The “Develop Pharmacophore Model” workflow imple-
mented in Phase47 involves two main steps. First, the “active
set” is used to generate several pharmacophore hypotheses, and a
score is assigned to each molecule of the set on the basis of the
alignment fitness of each active ligand onto the hypotheses.
Second, “inactive” molecules are aligned onto the hypotheses
and a score is assigned as described above for the “active set”. A
good hypothesis should have a high value for active scores and a
low value for inactive scores, i.e., a pharmacophore model should
be able to discriminate active molecules from inactive ones.
Initially, we found no pharmacophore models common to all
“active set” ligands. This most likely relates to the lack of
structural target information and clear knowledge about how
EPIs interact with NorA. Active molecules could indeed interact
with NorA using several different binding modes/sites. On the
basis of this assumption, the pharmacophore search was then
relaxed so that common pharmacophore hypotheses needed
only to match a subset of the chosen active compounds.
Common 3D models able to satisfy at least seven of 14 active
compounds and containing three, four, and five pharmacophore
sites were searched, obtaining several hypotheses which were
then scored and ranked. The analysis of the three and five site
models resulted in an inability to discriminate between active and
inactive molecules. As such, these pharmacophores were
rejected.
For four sites hypotheses, 30 models were generated by
Phase.47 The top ranked pharmacophore model had a high
“active score” coupled with a low “inactive score”, indicating a
good predictive power. This hypothesis was retained for further
analysis. The hypothesis consisted of four pharmacophoric
features: one hydrogen-bond acceptor (A3), one positive charge
(P7) and two aromatic rings (R9 and R10) (Figure 4). This
pharmacophore will be henceforth referred to as APRR.
Eight (1,41 2,41 3b, 3y, 6,41 astemizole,44 paroxetine,48 and
paroxetine-isomer48) of the 14 molecules belonging to the active
D
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
set perfectly matched the APRR hypothesis since they were able
to map simultaneously all the four pharmacophoric features in at
least one conformation (Table S-2 in Supporting Information).
The alignment of compound 1,41 which was the NorA EPI
showing the hightest fitness score value, and paroxetine48 onto
the APRR hypothesis is provided in Figure 5.
With respect to the “inactive set”, 49 out of the 51 compounds
were unable to map simultaneously to all four identified
pharmacophore features, highlighting the ability of the APRR
model to discriminate between active and inactive molecules.
Inactive ligands able to fit three out of the four sites always failed
to map the P7 feature, emphasizing the importance of a
protonable group for potent NorA inhibitory activity. For
instance, even though the inactive C-4 O-3-methylbutyl
derivative 3a is structurally related to the active quinoline 1,41
it is almost three times less active than the parent compound
(Table 1). The loss of the activity of compound 3a is obviously
due to the lack of the protonable side chain at the C-4 position of
the quinoline scaffold. Compounds 3t and 33 (Table 1 and Table
S-1 in Supporting Information, respectively) were the only
outliers, as they were able to match all the four sites of the APRR
pharmacophore model despite their lack of NorA inhibitory
activity.
To test whether this discrepancy was due to differences in
lipophilicity, the software QikProp52 was used to calculate the
descriptor QPlogPo/w, i.e., the predicted log of octanol/water
partition coefficient. The scatter plot reported in Figure 6
highlights how both compounds are outliers with respect to the
active compounds used for the model generation.
Figure 5. Superposition of the active quinoline derivative 141 (A) and
paroxetine48 (B) into the APRR model.
Figure 6. Scatter plot of QPlogPo/w for compounds 3t and 33 (black
labeled circles). Compounds belonging to the active set are plotted as
black crosses.
Compound 141 perfectly fit onto the four chemical
functionalities of the pharmacophore model. In particular, the
oxygen atom of the propoxy side chain occupied the hydrogen-
bond acceptor feature A3, the benzene ring of the quinoline
nucleus was disposed over the aromatic ring feature R10, the
phenyl substituent at the C-2 position overlapped with the
aromatic ring feature R9, and finally, the protonable amino
substituent at the C-4 position of the quinoline scaffold matched
the positively charged site P7. Considering paroxetine,48 the
hydrogen-bond acceptor feature A3 was mapped by one of the
oxygen atoms of the 1,3-benzodioxole nucleus, the aromatic sites
R9 and R10 were occupied by the two phenyl rings, and the
positively charged site P7 was matched by the protonable
nitrogen atom of the piperidine ring. It has been reported that
paroxetine inhibits the NorA efflux pump through competitive
mechanisms.48 Because paroxetine and quinoline derivatives
possessed a common pharmacophore, it is plausible to speculate
that our quinoline derivatives might inhibit NorA through the
same mechanism.
Thus, despite the presence of the four identified key
pharmacophoric features, compounds 3t and 33 could be
inactive as NorA inhibitors because of an unsatisfactory
pharmacokinetic profile.
PHARMACOPHORE-BASED VIRTUAL SCREENING
OF A FOCUSED LIBRARY
■
To aid the optimization of our 2-phenylquinoline EPIs, the
APRR pharmacophore model was used to rationally drive the
chemical synthesis of new 2-phenylquinoline derivatives. In
particular, the postulated APRR hypothesis was used to support
the selection of the most suitable protonable alkylamino chains
for our EPIs. Toward this end, a combinatorial focused virtual
library of 48.157 compounds was generated using the program
CombiGlide53 following the synthetic pathways described in
Figure 7, in which the quinoline synthons were reacted in silico
with the appropriate reagents retrieved from the Sigma-Aldrich
Web site.
E
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Chart 1. Side Chains Linked to the Hydroxyl Group at the C-4
Position of 2-Phenylquinolines
Figure 7. Schematic representation of the pathways for the synthesis of
the 2-phenylquinoline derivatives constituting the virtual library.
Table 3. 2-Phenylquinoline Derivatives Retrieved from the
Focused Library and Selected for Synthesis and Biologic
Evaluation
compd
fitness
compd
fitness
3d
R: 2.83
S: 2.80
R: 2.77
S: 2.66
R: 2.75
S: 2.12
R: 2.22
S: 1.89
2.81
3p
3q
3r
3s
2.20
3e
3f
2.53
2.71
2.58
3g
3h
3i
3u
3v
3x
3z
2.82
2.78
2.67
2.79
3.00
1.98
2.67
3k
3l
2.78
2.81
3m
3n
3o
2.61
141
paroxetine48
2.74
2.13
Compounds with a molecular weight greater than 500 Da and
more than two chiral centers were ruled out from the virtual
library, and the filtered structures (5060 molecules) were then
submitted to conformational sampling using the software
MacroModel.54 The generated conformations were used as
input to build a Phase database for the subsequent
pharmacophore screening. In fact, our aim was to use the
APRR hypothesis as a query to retrieve from the 5060
compounds library a set of quinoline derivatives containing the
specified 3D pattern of chemical functions. For this purpose, only
compounds with a fitness score greater than 2 and matching four
out of four sites (461 compounds) were retrieved and visually
inspected.
(propyloxy)phenyl]quinolin-4-ol (4)41 with 1-iodo-3-methylbu-
tane, 3-methanesulfonylalkylamines, 2-(chloromethyl)pyridine,
or the appropriate 2-chloroalkylamines, using K2CO3 in dry
DMF (Scheme 1). Reaction of 441 with 1-bromo-2-chloroethane
gave, under the same conditions, the intermediate 4-[(2-
chloroethyl)oxy]-2-[4′-(propyloxy)phenyl]quinoline (5),
which was further functionalized with the appropriate amines
using K2CO3 in dry DMF to obtain the 2-phenylquinoline
derivatives 3l−x. The hydrazinolysis of the phthalimide
derivative 3w, using NH2−NH2 and EtOH at reflux, resulted in
derivative 3y. The hydrolysis of the spiro derivative 3x with 9%
HCl solution at reflux, gave compound 3z.
Finally, a total of 19 2-phenylquinolines (3d−i, 3k−s, 3u, 3v,
3x, and 3z; see Chart 1 and Table 4 for the substituents) were
selected for synthesis mainly on the basis of their fitness value
(Table 3) and synthetic accessibility. These compounds were
then subjected to biologic evaluation.
To obtain compound 3c, the 1-(2-chloroethyl)pyrrole chain
has been synthesized following the procedure reported in
literature55 starting from 2,5-bis(methyloxy)tetrahydrofuran and
2-chloroethylamine hydrochloride.
After purification, compounds 3b, 3d, 3e, 3i, and 3o, obtained
as colorless oils, were converted to their corresponding solid HCl
salts, through bubbling HClg in their solutions in Et2O.
The reaction between key syntone 441 and (2-chloropropyl)-
dimethylamine is of particular interest (Scheme 2). After
CHEMISTRY
■
The C-4 O-alkyl derivative (3a) and O-alkyl amines (3b−k) were
prepared directly through the alkylation of key syntone 2-[4′-
F
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
a
Scheme 1
Table 4. EtBr Efflux Inhibition Activity (%) at 50 μM
Concentration, IC50, and MICs of the Second Set of 2-
Phenylquinoline Derivatives, against SA-1199B (norA
+/A116E GrlA)
a
Reagents and conditions: (i) ClCH2CH2Br, 1-iodo-3-methylbutane
or halo/mesylalkylamines, K2CO3, dry DMF, 70 °C, 11−65%; (ii)
phthalimide or cyclic/dialkylamines, K2CO3, dry DMF, 80 °C, 18−
54%; (iii) NH2−NH2, EtOH, reflux, 8 h, 14%; (iv) 9% HCl, MeOH,
reflux, 6 h, 40%.
a
Scheme 2
aziridinium ion to give compounds 3d and 3e, respectively. The
evidence that compound 3e was obtained in higher yield seems
to confirm our hypothesis about the mechanism of reaction,
given that the secondary carbon was less hindered than the
tertiary one.
a
Reagents and conditions: (i) K2CO3, dry DMF, 70 °C, 18−34%.
purification by flash column chromatography, two derivatives
were isolated and their H NMR spectra led to identify the
1
regioisomers 3d and 3e as racemates. This observation led us to
hypothesize the reaction mechanism: in basic conditions, the (2-
chloropropyl)dimethylamine rearranges to an aziridinium ion
intermediate, thus the potassium salt of compound 441 can react
with both the tertiary or the secondary carbon atom of the
RESULTS AND DISCUSSION
■
In this work, a first set of seven 2-phenylquinoline derivatives
bearing O-alkyl or O-alkylamines substituents at the C-4 position
that differed in size and degree of basicity were synthesized and
G
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 8. Effect of compounds 3e, 3f, 3h, 3i, 3m, 3q, 3s, 3x, 1,41 and reserpine49 on the MIC of CPX against S. aureus ATCC 25923 (wt), SA-K1902
(norA−), SA-K2378 (norA++), SA-1199 (norA wt), and SA-1199B (norA+/A116E GrlA).
tested. These results, together with data from literature, were
used to build a pharmacophore model for NorA EPIs. The model
was then used to design 19 additional compounds carrying
different substituents at C-4 position, which were in turn
synthesized and biologically evaluated.
Table 5. Intrinsic Antibacterial Activity (MICs, μg/mL) of
Compounds 3e, 3f, 3h, 3i, 3m, 3q, 3s, and 3x, 1,41 Reserpine,49
and CPX against the Five S. aureus Strains Used in the Test of
Synergism with CPX
MIC (μg/mL)
All the synthesized compounds were preliminarily screened
for their EtBr efflux inhibition activity, including in the test the
quinoline progenitor 141 and reserpine49 as reference com-
pounds (Tables 1 and 4).
SA-
K2378
(norA+
+)
SA-
1199
(norA
wt)
SA-1199B
(norA
+/A116E
GrlA)
S. aureus
SA-
ATCC
K1902
compd
3e
25923 (wt) (norA−)
nd
nd
nd
nd
nd
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
0.63
>100
>100
>100
>100
>100
>100
>100
>100
>100
>100
10
The EtBr efflux inhibitory activity of a 50 μM concentration of
some derivatives was superior to that of the reference compound
reserpine49 against SA-1199B (norA+/A116E GrlA). In
particular, compounds 3e, 3g, 3i, 3l, 3q, and 3r showed an
ability to inhibit EtBr efflux by more of the 90%, while derivatives
3d, 3f, 3h, 3k, 3m, 3o, 3s, 3u, and 3x had an inhibitory activity
comparable to that of reserpine49 (∼ 80%). Three additional
compounds (3n, 3v, and 3z) inhibited EtBr efflux by ∼70%,
while compound 3p showed lower inhibitory activity (57.7%).
Of note, compounds 3l and 3q showed a comparable or higher
inhibitory activity than the reference compound 141 (93.4%).
Dose−response EtBr efflux studies were performed for all
compounds inhibiting efflux by at least 80% (3d−i, 3k−m, 3q−s,
3f
nd
3h
>100
>100
>100
>100
>100
nd
100
25
>100
25
3i
3m
50
50
3q
25
50
3s
>100
nd
>100
nd
3x
141
reserpine49
CPX
100
50
50
>100
0.31
>100
0.63
>100
2.50
H
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
Figure 9. Effect of compounds 3h, 3i, 3m, 3q, and 3s on the MIC of EtBr against S. aureus SA-K2885 (mepA−) and SA-K2886 (mepA++).
3u, and 3x) (data not shown). These studies identified eight
compounds (3e, 3f, 3h, 3i, 3m, 3q, 3s, and 3x) having IC50 values
lower than 9 μM, lower than those of both reserpine49 and 141
(Table 4).
In an attempt to delineate a preliminary SAR for this emerging
class of EPIs, it is possible to state that: (i) the highest activities
were generally displayed by compounds having the 2-ethylamino
alkyl chains linked to the C-4 hydroxyl group (compounds 3d,
3e, and 3h) and (ii) the activity was maintained when the side
chain nitrogen atom was included in an aliphatic ring
(compounds 3f, 3m, 3q, 3s, and 3x). The influence of the
second heteroatom on the aliphatic ring moiety has not yet been
completely delineated. Inhibitory activity tended to be lost when
the second heteroatom was an oxygen or a sulfur atom. This was
confirmed by the fact that the piperazine derivative 3q displayed
EtBr efflux inhibitory activity of 93.4% in comparison to
morpholine 3k and thiomorpholine 3p, which inhibited the
efflux of 86.9% and 57.7%, respectively.
The MICs of the eight most active compounds (3e, 3f, 3h, 3i,
3m, 3q, 3s, and 3x) were determined using five S. aureus strains
with different norA genotypes. Data shown in Table 5 illustrate
how these selected compounds had little or no intrinsic
antibacterial activity, with MICs ranging from 25 to >100 μg/mL.
The compounds were then assayed for their ability to reduce
the MIC of CPX (synergistic activity) against the above-
mentioned S. aureus strains (Figure 8). For compounds 3e, 3f,
and 3x, synergistic activity data are available only for SA-1199
(norA wt) and SA-1199B (norA+/A116E GrlA).
Inhibitory activity was completely lost when the side chain
nitrogen was included in an aromatic ring (compounds 3c and
3j), in a phthalimide moiety (3w) or disubstituted with benzyl
groups (3t). This behavior may be related to the reduction of the
nitrogen nucleophilicity, as supported by the low inhibitory
activity of the compound 3a, which is missing a protonable
nitrogen, or to steric hindrance. The monosubstitution with a
benzyl group was, on the other hand, well tolerated. In fact, the
compound 3v, in which the side chain nitrogen atom is
substituted with an ethyl and a benzyl group, possessed an
EtBr inhibitory activity higher than that of the compound 3t,
having two benzyl groups as substituents.
In conclusion, chemical modifications proposed by in silico
analyses led us to confirm the 2-(4′-propoxyphenyl)quinoline
nucleus as a very good scaffold to obtain new and potent EPIs.
Some of the 2-{[2-(4′-propoxyphenyl)-4-quinolinyl]-oxy}-
ethanamine derivatives, showing modest or no intrinsic
antistaphylococcal activity up to the highest concentration tested
(100 μg/mL), were able to restore, in a concentration-dependent
manner, the antibacterial activity of CPX against norA-over-
expressing S. aureus strains. Both the EPI activity of these
derivatives in the EtBr efflux inhibition assay and their synergistic
activity with CPX were superior to that shown by the reference
compounds against the norA-overexpressing strain SA-1199B.
Compounds 3m and 3q also possessed potent synergistic activity
with EtBr against a MepA overexpressing strain.
The isobolograms illustrated in Figure 8 revealed that neither
the tested compounds nor the reference compound reserpine49
showed any significant synergistic activity with CPX against S.
aureus ATCC 25923 (wt), SA-1199 (norA wt), and SA-K1902
(norA−) strains. These data are in agreement with the low
expression of the NorA pump in these strains. On the contrary, a
good synergistic effect with CPX was demonstrated by all the
tested compounds against SA-K2378 (norA++), with a CPX
MIC reduction of 4−32-fold (2.5 → 0.08 μg/mL). In particular,
0.78 μg/mL of compound 3m reduced the CPX MIC 16-fold.
When tested against SA-1199B (norA+/A116E GrlA), 3.13 μg/
mL of compounds 3m and 3q showed the best synergistic activity
by reducing the CPX MIC 16-fold (10 → 0.63 μg/mL).
Given their potent NorA inhibitory activity, the inhibitory
activities of compounds 3h, 3i, 3m, 3q, and 3s were evaluated
against the MepA (MATE superfamily) efflux pump. This was
accomplished using S. aureus strains SA-K2885 (mepA−) and
SA-K2886 (mepA++, overexpressed from a plasmid) (Figure 9).
In both cases, synergistic activity studies were carried out using
EtBr instead of CPX, as it is a better MepA substrate and thus
permits a clearer picture of MepA inhibition.
As expected, the tested compounds did not show any
synergistic activity with EtBr against SA-K2885 (mepA−).
However, all were able to reduce the EtBr MIC against SA-
K2886 (mepA++). The most active compounds were 3m and 3q,
which were able to reduce the EtBr MIC 16-fold at
concentrations of 0.78 and 1.56 μg/mL, respectively. These
data indicate that 2-(4′-propoxyphenyl)]quinolines 3h, 3i, 3m,
3q, and 3s are inhibitors of both the NorA and MepA MDR efflux
pumps and are able to completely restore the activity of CPX and
EtBr against strains overexpressing such pumps.
EXPERIMENTAL SECTION
■
Preparation of the Data Set Compounds. Molecules were built
using the fragment library tools of the Maestro GUI56 and then
submitted to conformational search using the software MacroModel.54
I
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
The search was carried on in vacuum using the OPLS 2005 force field.
The “Mixed torsional/Low mode sampling method” was employed,
using the automatic setup of the search variables; the maximum numbers
of steps were set to 10000 and 100 per molecule and rotatable bond,
respectively. Conformers in an energy window of 10 kcal/mol were
saved, discarding the ones redundant on the basis of their RMSD.
Minimization of conformers was performed using the Polak−Ribiere
conjugate gradient method, using a maximum of 500 minimization step
and 0.0005 kJ/(Å mol) gradient convergence threshold.
rotary evaporator under reduced pressure. All starting materials were
commercially available, unless otherwise indicated.
General Procedure for the Synthesis of the 4-[(2-
Chloroethyl)oxy]-2-[4′-(propyloxy)phenyl]quinolone (5), 4-[(3-
Methylbutyl)oxy]-2-[4′-(propyloxy)phenyl]quinoline (3a), and
the [2-({2-[4′-(Propyloxy)phenyl]quinolin-4-yl}oxy)alkyl]-
amines (3b−k). To a solution of 2-[4′-(propyloxy)phenyl]quinolin-
4-ol 441(1.0 equiv) in dry DMF (10 mL) was added K2CO3 (4.0 equiv)
and a solution of 1-bromo-2-chloroethane or 1-iodo-3-methylbutane or
halo/mesylalkylamines (4.0 equiv) in dry DMF (3 mL) was added
dropwise. The reaction mixture was warmed to 70 °C under nitrogen
atmosphere with stirring until 2-[4′-(propyloxy)phenyl]quinolin-4-ol
441 disappeared (TLC). The mixture was then poured in water and
extracted with EtOAc (3 × 100 mL), and the combined organic extracts
were dried over Na2SO4. The organic solvent was then removed under
reduced pressure to obtain a residue that was purified by flash column
chromatography to give the target compounds. For compounds 3b, 3d,
3e, and 3i, the obtained oil was dissolved in a mixture of Et2O/EtOH
and treated with HClg until the target compound precipitated as its HCl
salt.
4-[(2-Chloroethyl)oxy]-2-[4′-(propyloxy)phenyl]quinolone (5).
Obtained from 1-bromo-2-chloroethane as a white solid after
purification by flash chromatography (silica gel/PE/EtOAc 95/5)
(65% yield, mp 100.3−101.8 °C). 1H NMR (CDCl3): δ 1.11 (3H, t, J =
7.35 Hz, OCH2CH2CH3), 1.84−1.95 (2H, m, OCH2CH2CH3), 4.02−
4.08 (4H, m, OCH2CH2CH3,OCH2CH2Cl), 4.59 (2H, t, J = 5.67 Hz,
OCH2CH2Cl), 7.07 (2H, d, J = 8.87 Hz, H-3′ and H-5′), 7.16 (1H, s, H-
3), 7.52 (1H, t, J = 7.53 Hz, H-6), 7.75 (1H, t, J = 5.24 Hz, H-7), 8.10−
8.15 (3H, m, H-8, H-2′ and H-6′), 8.26 (1H, d, J = 8.26 Hz, H-5).
4-(3-Methylbutoxy)-2-(4′-propoxyphenyl)quinoline (3a). Ob-
tained from 1-iodo-3-methylbutane as a yellowish solid after purification
by flash chromatography (silica gel/PE/EtOAc 97/3) (40% yield, mp
78.2−83.1 °C). 1H NMR (CDCl3): δ 0.98−1.18 (9H, m, CH3), 1.98−
2.07 (5H, m, OCH2CH2CH, OCH2CH2CH and OCH2CH2CH3), 4.04
(2H, t, J = 6.62 Hz, OCH2CH2CH3), 4.34 (2H, t, J = 6.21 Hz,
OCH2CH2CH), 7.08 (2H, d, J = 8.35 Hz, H-3′ and H-5′), 7.17 (1H, s,
H-3), 7.50 (1H, t, J = 7.08 Hz, H-6), 7.73 (1H, t, J = 7.05 Hz, H-7),
8.10−8.14 (3H, m, H-2′, H-6′ and H-8), 8.23 (1H, d, J = 8.21 Hz, H-5).
Anal. (C23H27NO2) C, H, N.
N,N-Dimethyl-2-{[2-(4′-propoxyphenyl)-4-quinolinyl]oxy}-
ethanamine Hydrochloride (3b). Obtained from (2-chloroethyl)-
dimethylamine hydrochloride as a white solid after purification by flash
chromatography (silica gel/CH2Cl2/MeOH 99/1) and precipitation as
hydrochloride (56% yield, mp 145.1−146.9 °C). 1H NMR (DMSO-d6):
δ 0.95 (3H, t, J = 7.44 Hz, OCH2CH2CH3), 1.68−1.78 (2H, m,
OCH2CH2CH3), 2.85 (6H, d, J = 4.72 Hz, NCH3), 3.65−3.80 (2H, m,
OCH2CH2N), 4.04 (2H, t, J = 6.49 Hz, OCH2CH2CH3), 4.90−5.20
(2H, m, OCH2CH2N), 7.20 (2H, d, J = 8.74 Hz, H-3′ and H-5′), 7.71−
7.79 (2H, m, H-3 and H-6), 8.02 (1H, t, J = 8.11 Hz, H-7), 8.86 (2H, d, J
= 8.86 Hz, H-2′and H-6′), 8.50 (1H,d, J = 8.32 Hz, H-8), 8.59 (1H, d, J =
8.30 Hz, H-5), 11.16 (1H, bs, NH). Anal. (C22H26N2O2·HCl) C, H, N.
2-(4′-Propoxyphenyl)-4-[2-(1H-pyrrol-1-yl)ethoxy]quinoline (3c).
Obtained from 1-(2-chloroethyl)-1H-pyrrole55 as a white solid after
purification by flash chromatography (silica gel/PE/EtOAc 95/5) (30%
yield, mp 127.7−129.0 °C). 1H NMR (CDCl3): δ 1.11 (3H, t, J = 7.35
Hz, OCH2CH2CH3), 1.84−1.94 (2H, m, OCH2CH2CH3), 4.04 (2H, t, J
= 6.61 Hz, OCH2CH2N), 4.53 (4H, m, OCH2CH2N and
OCH2CH2CH3);6.25 (2H, d, J = 2.11 Hz, pyrrole CH), 6.89 (2H, t, J
= 2.10 Hz, pyrrole NCH), 7.04−7.09 (3H, m, H-3, H-3′ and H-5′), 7.52
(1H, t, J = 2.55 Hz, H-6), 7.73 (1H, t. J = 2.57 Hz, H-7), 8.05−8.13 (3H,
m, H-8, H-2′ and H-6′), 8.20 (1H, d, J = 7.20 Hz, H-5). Anal.
(C24H24N2O2) C, H, N.
Pharmacophore Model Generation. Phase 3.347 was used to
carry on the pharmacophore modeling studies. First, we established an
arbitrary range of activities for all the molecules in the data set (activity is
expressed as percentage of inhibition of EtBr efflux). Molecules showing
a percentage of EtBr efflux inhibition greater than 70% were considered
“active”, while molecules showing a percentage of less than 30% were
considered “inactive”. The “active set” was then composed of 14
inhibitors (Table 2), whereas the “inactive” one was composed of 51
molecules (Table S-1 in Supporting Information).
Pharmacophoric sites were generated using the default feature
definitions as implemented in Phase. Pharmacophore features common
to at least seven of 14 active compounds and composed by three, four or
five sites were generated, obtaining several hypotheses. The hypotheses
were scored by the goodness of the alignment of each active ligand onto
hypotheses (“active score”). The scoring algorithm took into account
contributions from the alignment of site points and vectors, volume
overlap, selectivity, number of active ligands matched, relative
conformational energy, and ligand activity. A second score (“inactive
score”) was then calculated using the alignment of each inactive
molecule. A satisfying hypothesis should show high active scores as well
as low inactive scores, thus proving to be able to discriminate between
active and inactive molecules.
Following the active/inactive scoring ratio as guideline, the best
identified hypothesis (APRR: active score = 5.038, inactive score =
1.224) was composed by four pharmacophoric features: one hydrogen-
bond acceptor (A3), one positive ionizable group (P7), and two
aromatic rings (R9 and R10).
Focused Library Generation. The Combinatorial Library
Enumeration utility in CombiGlide57 was used to build a focused
library of quinoline derivatives. Following the a priori decided two
synthetic pathways (Figure 7), cores were sketched using the Maestro
GUI, the appropriate reagents were searched through the Sigma-Aldrich
Web site, downloaded as an sdf file, and prepared with the Reagent
Preparation utility in CombiGlide. Cores were then reacted in silico with
the reactants to obtain all the possible combinations for each reaction,
and structures were untangled and minimized. Output structures with a
molecular weight greater than 500 Da and more than two chiral centers
were ruled out, and the filtered structures were then submitted to a
conformational sampling using MacroModel with the same settings
used for the data set. Finally, the conformations were used as input to
create a phase database (5060 molecules).
Synthesis. All reactions were routinely checked by thin-layer
chromatography (TLC) on silica gel 60F254 (Merck) and visualized
using UV illumination. Flash chromatography was performed on Merck
silica gel 60 (mesh 230−400) using the indicated solvents. Yields were of
purified product and were not optimized. Melting points were
determined in capillary tubes (Mettler PF62 apparatus) and are
uncorrected. Elemental analyses were performed by a Fisons elemental
analyzer (model EA1108CHN), and the data for C, H, and N are within
1
0.4% of the theoretical values (≥95% purity). H NMR spectra were
recorded at 400 MHz with a Bruker Advance-DRX 400 instrument and
with Me4Si as the internal standard. The chemical shift (δ) values are
reported in ppm, and the coupling constants (J) are given in Hz. The
abbreviations used are as follows: s, singlet; bs, broad singlet; d, doublet;
dd, double doublet; t, triplet; m, multiplet. The spectral data are
consistent with the assigned structures. Reagents and solvents were
purchased from common commercial suppliers and were used as
received. For routine aqueous workup, the reaction mixture was
extracted with CH2Cl2 or EtOAc. The organic layer was washed with
N,N-Dimethyl-2-{[2-(4′-propoxyphenyl)quinolin-4-yl]oxy}propan-
1-amine Hydrochloride (3d) and N,N-Dimethyl-1-{[2-(4′-propox-
yphenyl)-4-quinolinyl]oxy}-2-propanamine Hydrochloride (3e). Ob-
tained from (2-chloropropyl)dimethylamine hydrochloride after
purification by flash chromatography (silica gel/CH2Cl2/EtOH 97/3)
and precipitation as its HCl salt.
1
3d: 18% yield; mp 189.7−190.5 °C. H NMR (DMSO-d6): δ 1.04
brine, dried over anhydrous Na SO , and concentrated with a Buchi
(3H, t, J = 7.33 Hz, OCH2CH2CH3), 1.47 (3H, d, J = 5.89 Hz,
̈
2
4
J
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
OCHCH3CH2N), 1.76- 1.87 (2H, m, OCH2CH2CH3), 2.80−2.95 (8H,
m, OCHCH3CH2N and NCH3), 3.75−3.85 (1H, m, OCHCH3CH2N),
4.07 (2H, t, J = 5.03 Hz, OCH2CH2CH3), 7.17 (2H, d, J = 9.09 Hz, H-3′
and H-5′), 7.63 (1H, t, J = 8.43 Hz, H-6), 7.73 (1H, s, H-3), 7.87 (1H, t, J
= 8.51 Hz, H-7), 8.25 (1H, d, J = 8.39 Hz, H-8), 8.50−8.75 (3H, m, H-5,
H-2′ and H-6′), 10.40 (1H, bs, NH). Anal. (C23H28N2O2·HCl) C, H, N.
3e: 34% yield; mp 80.7−81.5 °C. 1H NMR (DMSO-d6): δ 1.05 (3H,
t, J = 7.21 Hz, OCH2CH2CH3), 1.54 (3H, d, J = 6.66 Hz,
OCH2CHCH3N), 1.77−1.79 (2H, m, OCH2CH2CH3), 2.90 (6H, d, J
= 4.66 Hz, NCH3), 4.05−4.20 (3H, m,OCH2CH2CH3 and
OCH2CHCH3N), 4.95 (2H, d, J = 4.61 Hz, OCH2CHCH3N), 7.29
(2H, d, J = 8.84 Hz, H-3′ and H-5′), 7.79−7.87 (2H, m, H-3 and H-6),
8.05 (1H, t, J = 8.70 Hz, H-7), 8.32 (2H, d, J = 8.93 Hz, H-2′ and H-6′),
8.51 (1H, d, J = 8.51 Hz, H-8), 8.63 (1H, d, J = 8.77 Hz, H-5), 11.10 (1H,
bs, NH). Anal. (C23H28N2O2·HCl) C, H, N.
8.81 Hz, H-3′and H-5′), 7.22−7.35 (2H, m, H-3 and pyridine CH), 7.54
(1H, t, J = 6.99 Hz, H-6), 7.66−7.85 (3H, m, H-7 and pyridine CH),
8.05−8.15 (3H, m, H-8, H-2′ and H-6′), 8.34 (1H, d, J = 7.52 Hz, H-8),
8.70 (1H, d, J = 4.75 Hz, pyridine CH). Anal. (C24H22N2O2) C, H, N.
4-[2-(4-Morpholinyl)ethoxy]-2-(4′-propoxyphenyl)quinoline (3k).
Obtained from 4-(2-chloroethyl)morpholine hydrochloride as a white
solid after purification by flash chromatography (silica gel/CH2Cl2/
MeOH 99/1) (24% yield, mp 141.5−142.5 °C). 1H NMR (CDCl3): δ
1.02 (3H, t, J = 7.44 Hz, OCH2CH2CH3), 1.71−1.89 (2H, m,
OCH2CH2CH3), 2.61−2.73 (4H, m, morpholine NCH2), 3.05 (2H, t, J
= 4.99 Hz, OCH2CH2N), 3.72−3.83 (4H, m, morpholine OCH2), 3.95
(2H, t, J = 6.57 Hz, OCH2CH2CH3), 4.48 (2H, t, J = 5.33 Hz,
OCH2CH2N), 6.98 (2H, d, J = 8.75 Hz, H-3′ and H-5′), 7.12 (1H, s, H-
3), 7.44 (1H, t, J = 7.98 Hz, H-6), 7.68 (1H, t, J = 7.73 Hz, H-7), 8.03−
8.10 (3H, m, H-8, H-2′ and H-6′), 8.21 (1H, d, J = 8.16 Hz, H-5). Anal.
(C24H28N2O3) C, H, N.
4-[(1-Methylpiperidin-3-yl)oxy]-2-(4′-propoxyphenyl)quinoline
(3f). Obtained from 1-methylpiperidin-3-yl methanesulfonate58 as a
white solid after purification by flash chromatography (silica gel/
CH2Cl2/MeOH 99/1) (16% yield, mp 72.5−73.8 °C). 1H NMR
(CDCl3): δ 1.11 (3H, t, J = 7.34 Hz, OCH2CH2CH3), 1.84−1.94 (6H,
m, OCH2CH2CH3 and piperidine CH2), 2.08−2.36 (2H, m, piperidine
CH2), 2.63 (3H, s, NCH3), 2.90−3.30 (2H, m, piperidine CH2), 4.05
(2H, t, J = 6.60 Hz, OCH2CH2CH3), 4.05 (2H, t, J = 6.60 Hz,
OCH2CH2CH3), 4.30−4.35 (1H, m, piperidine CH), 7.07 (2H, d, J =
8.89 Hz, H-3′ and H-5′), 7.31 (1H, s, H-3), 7.52 (1H, t, J = 8.15 Hz, H-
6), 7.72 (1H, t, J = 8.08 Hz, H-7), 8.09−8.13 (3H, m, H-8, H-2′ and H-
6′), 8.22 (1H, d, J = 8.29 Hz, H-5). Anal. (C25H30N2O2) C, H, N.
4-(1-Azabicyclo[2.2.2]oct-3-yloxy)-2-(4′-propoxyphenyl)quinoline
(3g). Obtained from 3-chloro-1-azabicyclo[2.2.2]octane hydrochloride
as a colorless oil after purification by flash chromatography (silica gel/
CH2Cl2/EtOH 98/2) (11% yield). 1H NMR (CDCl3): δ 1.08 (3H, t, J =
7.36 Hz, OCH2CH2CH3), 1.63−1.92 (6H, m, OCH2CH2CH3 and 2 ×
CH2 azabicyclo), 2.41−2.53 (1H, m, CH azabicyclo)), 2.88−3.51 (6H,
m, 3 × CH2 azabicyclo), 4.02 (2H, t, J = 6.63 Hz, OCH2CH2CH3),
4.74−4.78 (1H, m, CH azabicyclo), 6.95−7.08 (3H, m, H-3, H-3′ and
H-5′), 7.47 (1H, t, J = 6.70 Hz, H-6), 7.70 (1H, t, J = 7.05 Hz, H-7),
8.03−8.10 (3H, m, H-8, H-2′ and H-6′), 8.21 (1H, d, J = 7.05 Hz, H-5).
Anal. (C25H28N2O2) C, H, N.
N,N-Diisopropyl-N-(2-{[2-(4′-propoxyphenyl)-4-quinolinyl]oxy}-
ethyl)amine (3h). Obtained from N-(2-chloroethyl)-N-(1-
methylethyl)propan-2-amine hydrochloride as a white solid after
purification by flash chromatography (silica gel/PE/EtOAc 80/20)
and precipitated as its HCl salt (25% yield, mp 75.5−78.8 °C). 1H NMR
(CDCl3): δ 1.10−1.13 (15H, m, OCH2CH2CH3 and isopropylic CH3),
1.81−1.91 (2H, m, OCH2CH2CH3), 3.00−3.16 (4H, m, OCH2CH2N
and isopropylic CH), 4.01 (2H, t, J = 6.61 Hz, OCH2CH2CH3), 4.20
(2H, t, J = 7.17 Hz, OCH2CH2N), 7.04 (2H, d, J = 8.80 Hz, H-3′ and H-
5′), 7.17 (1H, s, H-3), 7.46 (1H, t, J = 8.09 Hz, H-6), 7.69 (1H, d, J = 8.05
Hz, H-7), 8.07−8.11 (3H, m, H-8, H-2′ and H-6′), 8.20 (1H, d, J = 8.26
Hz, H-5). Anal. (C26H34N2O2) C, H, N.
N,N-Dimethyl-3-{[2-(4′-propoxyphenyl)-4-quinolinyl]oxy}-1-prop-
anamine Hydrochloride (3i). Obtained from (3-chloropropyl)-
dimethylamine hydrochloride as a white solid after purification by
flash column chromatography (silica gel/CHCl3/MeOH 99/1) and
precipitated as its HCl salt (57% yield, mp 145.1−146.2 °C). 1H NMR
(DMSO-d6): δ 1.04 (3H, t, J = 7.44 Hz, OCH2CH2CH3), 1.80−2.00
(2H, m, OCH2CH2CH3), 2.40−2.50 (2H, m, OCH2CH2CH2NCH3),
2.38 (6H, d, J = 4.85 Hz,OCH2CH2CH2N(CH3)2), 3.35−3.45 (2H, m,
OCH2CH2CH2N(CH3)2), 4.13 (2H, t, J = 6.57 Hz, OCH2CH2CH3),
4.81 (2H, t, J = 6.58 Hz, OCH2CH2CH2N(CH3)2), 7.28 (2H, d, J = 8.60
Hz, H-3′and H-5′), 7.79 (1H, s, H-3), 7.86 (1H, t, J = 8.16 Hz, H-6),
8.13 (1H, t, J = 7.00 Hz, H-7), 8.35 (2H, d, J = 8.61 Hz, H-2′ and H-6′),
8.45 (1H, d, J = 8.19 Hz, H-8), 8.68 (1H, d, J = 9.06 Hz, H-5). Anal.
(C23H28N2O2·HCl) C, H, N.
2-(4′-Propoxyphenyl)-4-(2-pyridinylmethoxy)quinoline (3j). Ob-
tained from 2-(chloromethyl)pyridine hydrochloride as a white solid
after purification by flash chromatography (silica gel/PE/EtOAc 90/10)
(62% yield, mp 138.0−138.8 °C). 1H NMR (CDCl3): δ 1.10 (3H, t, J =
7.30 Hz, OCH2CH2CH3), 1.84−1.94 (2H, m, OCH2CH2CH3), 4.03
(2H, t, J = 6.56 Hz, OCH2CH2N), 5.57 (2H, s, OCH2), 7.06 (2H, d, J =
General Procedure for the Synthesis of the [2-({2-[4′-
(Propyloxy)phenyl]quinolin-4-yl}oxy)alkyl]amines (3l−x). To a
solution of 4-[(2-chloroethyl)oxy]-2-[4′-(propyloxy)phenyl]quinolone
(5) (1.0 equiv) in dry DMF (10 mL) was added K2CO3 (4.0 equiv)
followed by the dropwise addition of a solution of the suitable cyclic/
dialkylamine (4.0 equiv) in dry DMF (3 mL). The reaction mixture was
warmed at 80 °C under a nitrogen atmosphere with stirring until 4-[(2-
chloroethyl)oxy]-2-[4′-(propyloxy)phenyl]quinolone (5) disappeared
(TLC). The mixture was then poured into water and extracted with
EtOAc (3 × 100 mL). The organic solvent was removed under reduced
pressure to obtain a residue that was purified by flash chromatography or
crystallized from cyclohexane to give the target compounds. For
compound 3o, the obtained oil was dissolved in a mixture of Et2O/
EtOH and treated with HClg until the target compound precipitated as
its HCl salt.
2-(4′-Propoxyphenyl)-4-[2-(1-pyrrolidinyl)ethoxy]quinoline (3l).
Obtained from pyrrolidine as a white solid after crystallization from
cyclohexane (18% yield, mp 99.7−101.2 °C). 1H NMR (CDCl3): δ 1.01
(3H, t, J = 7.47 Hz OCH2CH2CH3), 1.70−1.95 (6H, m, OCH2CH2CH3
and pyrrolidine CH2), 2.60−2.80 (4H, m, pyrrolidine NCH2), 3.06 (2H,
t, J = 5.86 Hz, OCH2CH2N), 3.96 (2H, t, J = 6.62 Hz, OCH2CH2CH3),
4.39 (2H, t, J = 5.85 Hz, OCH2CH2N), 6.97 (2H, d, J = 9.55 Hz, H-3′and
H-5′), 7.10 (1H, s, H-3), 7.39 (1H, t, J = 6.88 Hz, H-6), 7.63 (1H, t, J =
6.92, H-7), 7.95−8.05 (3H, m, H-8, H-2′and H-6′), 8.15 (1H, d, J = 8.26
Hz, H-5). Anal. (C24H28N2O2) C, H, N.
4-[2-(1-Azepanyl)ethoxy]-2-(4′-propoxyphenyl)quinoline (3m).
Obtained from azepane as a white solid after purification by flash
chromatography (silica gel/CH2Cl2/MeOH 99/1) (21% yield, mp
74.9−75.8 °C). 1H NMR (CDCl3): δ 1.11 (3H, t, J = 7.43 Hz,
OCH2CH2CH3), 1.55−2.10 (10H, m, OCH2CH2CH3 and azepane
CH2), 2.92−3.08 (4H, m, azepane NCH2), 3.26 (2H, t, J = 5.74 Hz,
OCH2CH2N), 4.05 (2H, t, J = 6.60 Hz, OCH2CH2CH3), 4.47 (2H, t, J =
5,90 Hz, OCH2CH2N), 7.08 (2H, d, J = 8.82 Hz, H-3′and H-5′), 7.21
(1H, s, H-3), 7.49 (1H, t, J = 7.03 Hz, H-6), 7.73 (1H, t, J = 8.35 Hz, H-
7), 8.05−8.15 (3H, m, H-8, H-2′, and H-6′), 8.19 (1H, d, J = 8.39 Hz, H-
5). Anal. (C26H32N2O2) C, H, N.
4-[2-(3,4-Dihydroisoquinolin-2(1H)-yl)ethoxy]-2-(4′-
propoxyphenyl)quinoline (3n). Obtained from 1,2,3,4-tetrahydroiso-
quinoline as a white solid after purification by flash chromatography
(silica gel/PE/EtOAc 90/10) (23% yield, mp 74.9−76.3 °C). 1H NMR
(CDCl3): δ 1.08 (3H, t, J = 7.41 Hz, OCH2CH2CH3) 1.82−1.95 (2H, m,
OCH2CH2CH3), 2.99 (4H, s, dihydroisoquinoline CH2), 3.20 (2H, t, J
= 5.77 Hz, OCH2CH2N), 3.89 (2H, s, dihydroisoquinoline CH2), 4.01
(2H, t, J = 6.59 Hz, OCH2CH2CH3), 4.52 (2H, t, J = 5.80 Hz,
OCH2CH2N), 7.03−7.07 (3H, m, dihydroisoquinoline CH, H-3′ and
H-5′), 7.13−7.21 (3H, m, dihydroisoquinoline CH), 7.27 (1H, s, H-3),
7.49 (1H, t, J = 7.95 Hz, H-6), 7.71 (1H, t, J = 7.01 Hz, H-7), 8.07−8.10
(3H, m, H-8, H-2′ and H-6′), 8.21 (1H, d, J = 8.15 Hz, H-5). Anal.
(C29H30N2O2) C, H, N.
4-[2-(6,7-Dimethoxy-3,4-dihydro-2(1H)-isoquinolinyl)ethoxy]-2-
(4′-propoxyphenyl)quinoline Hydrochloride (3o). Obtained from 6,7-
bis(methyloxy)-1,2,3,4-tetrahydroisoquinoline hydrochloride as a white
solid after purification by flash chromatography (silica gel/EtOAc/
1
EtOH 98/2) (68% yield, mp 138.7−140.2 °C). H NMR (CDCl3): δ
K
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
1.11 (3H, t, J = 7.31 Hz, OCH2CH2CH3), 1.80−2.00 (2H, m,
OCH2CH2CH3), 2.85−3.10 (4H, m, dihydroisoquinoline CH2), 3.23
(2H, t, J = 5.47 Hz, OCH2CH2N), 3.84 (2H, s, dihydroisoquinoline
CH2), 3.88 (3H, s, OCH3), 3.90 (3H, s, OCH3), 4.05 (2H, t, J = 6.62 Hz,
OCH2CH2CH3), 4.56 (2H, t, J = 5.74 Hz, OCH2CH2N), 6.58 (1H, s,
dihydroisoquinoline CH), 6.66 (1H, s, dihydroisoquinoline CH), 7.06
(2H, d, J = 8.85 Hz, H-3′and H-5′), 7.24 (1H, s, H-3), 7.51 (1H, t, J =
8.09 Hz, H-6), 7.74 (1H, t, J = 6.76 Hz, H-7), 8.03−8.18 (3H, m, H-8, H-
2′ and H-6′), 8.24 (1H, d, J = 8.26 Hz, H-5). Anal. (C31H34N2O4·HCl)
C, H, N.
1.60 (8H, m, 2 × NCH2CH2CH2CH3), 1.84−1.95 (2H, m,
OCH2CH2CH3), 2.64 (4H, t, J = 6.91 Hz, 2 × NCH2CH2CH2CH3),
3.10 (2H, t, J = 5.44 Hz, OCH2CH2N), 4.05 (2H, t, J = 6.07 Hz,
OCH2CH2CH3), 4.37 (2H, t, J = 6.11 Hz, OCH2CH2N), 7.07 (2H, d, J
= 8.81 Hz, H-3′ and H-5′), 7.19 (1H, s, H-3), 7.48 (1H, t, J = 7.57 Hz, H-
6), 7.72 (1H, t, J = 7.79 Hz, H-7), 8.08−8.15 (3H, m, H-8, H-2′ and H-
6′), 8.21 (1H, d, J = 8.28 Hz, H-5). Anal. (C28H38N2O2) C, H, N.
N-Benzyl-N-ethyl-2-{[2-(4′-propoxyphenyl)quinolin-4-yl]oxy}-
ethanamine (3v). Obtained from ethyl(phenylmethyl)amine as a white
semisolid after purification by flash chromatography (silica gel/CH2Cl2/
Et2O 97/3) (19% yield). 1H NMR (CDCl3): δ 1.11 (3H, t, J = 7.46 Hz,
OCH2CH2CH3), 1.20 (3H, t, J = 6.99 Hz, NCH2CH3), 1.85−1.95 (2H,
m, OCH2CH2CH3), 2.79 (2H, q, J = 7.00 Hz, NCH2CH3), 3.13 (2H, t, J
= 5.89 Hz, OCH2CH2N), 3.81 (2H, s, CH2Ph), 4.06 (2H, t, J = 6.59 Hz,
OCH2CH2CH3), 4.34 (2H, t, J = 6.05 Hz, OCH2CH2N), 7.05−7.11
(3H, m, H-3, H-3′ and H-5′), 7.30−7.47 (6H, m, H-6 and phenyl CH),
7.71 (1H, t, J = 7.77 Hz, H-7), 8.07−8.16 (4H, m, H-5, H-8, H-2′ and H-
6′). Anal. (C29H32N2O2) C, H, N.
2-[4′-(Propyloxy)phenyl]-4-[(2-thiomorpholin-4-ylethyl)oxy]-
quinoline (3p). Obtained from thiomorpholine as yellow a solid after
purification by flash chromatography (silica gel/CH2Cl2/MeOH 98/2)
(30% yield, mp 0.151.2−152.0 °C). 1H NMR (CDCl3): δ 1.11 (3H, t, J =
7.34 Hz, OCH2CH2CH3), 1.83−1.95 (2H, m, OCH2CH2CH3), 2.70−
2.80 (4H, m, thiomorpholine NCH2), 2.97−3.05 (4H, m, thiomorpho-
line SCH2), 3.10 (2H, t, J = 5.74 Hz, OCH2CH2N), 4.05 (2H, t, J = 6.63
Hz, OCH2CH2CH3), 4.43 (2H, t, J = 5.56 Hz, OCH2CH2N), 7.07 (2H,
d, J = 8.84 Hz, H-3′ and H-5′), 7.18 (1H, s, H-3), 7.50 (1H, t, J = 8.29 Hz,
H-6), 7.74 (1H, t, J = 8.18 Hz, H-7), 8.09−8.25 (4H, m, H-5, H-8, H-2′
and H-6′). Anal. (C24H28N2O2S) C, H, N.
2-(2-{[2-(4′-Propoxyphenyl)-4-quinolinyl]oxy}ethyl)-1H-isoindole-
1,3(2H)-dione (3w). Obtained from phthalimide potassium salt as a
white solid after purification by flash chromatography (silica gel/
cyclohexane/EtOAc 95/5) (54% yield, mp 195.5−196.3 °C). 1H NMR
(CDCl3): δ 1.11 (3H, t, J = 7.48 Hz, OCH2CH2CH3), 1.80−2.00 (2H,
m, OCH2CH2CH3), 4.07 (2H, t, J = 5.51 Hz, OCH2CH2N), 4.38 (2H, t,
J = 4.97 Hz, OCH2CH2CH3), 4.67 (2H, t, J = 5.97 Hz, OCH2CH2N),
7.07 (2H, d, J = 8.85 Hz, H-3′and H-5′), 7.18 (1H, s, H-3), 7.57 (1H, t, J
= 7.06 Hz, H-6), 7.70−7.87 (3H, m, H-8, H-5″ and H-6″), 7.90−8.05
(2H, m, H-4″ and H-7″), 8.10−8.25 (4H, m, H-5, H-7, H-2′ and H-6′).
Anal. (C28H24N2O4) C, H, N.
4-[2-(1,4-Dioxa-8-azaspiro[4.5]dec-8-yl)ethoxy]-2-(4′-
propoxyphenyl)quinoline (3x). Obtained from 1,4-dioxa-8-
azaspiro[4.5]decane as a white solid after purification by flash
chromatography (silica gel/PE/EtOAc 70/30) (45% yield, mp 125.0−
126.4 °C). 1H NMR (CDCl3): δ 1.10 (3H, t, J = 7.34 Hz,
OCH2CH2CH3), 1.81−1.94 (6H, m, OCH2CH2CH3, and piperidine
CH2), 2.76−2.85 (4H, m, piperidine CH2N), 3.09 (2H, t, J = 5.69 Hz,
OCH2CH2N), 4.00−4.08 (6H, m, OCH2CH2CH3 and dioxolane CH2),
4.45 (2H, t, J = 5.79 Hz, OCH2CH2N), 7.07 (2H, d, J = 8.83 Hz, H-3′
and H-5′), 7.18 (1H, s, H-3), 7.48 (1H, t, J = 6.98 Hz, H-6), 7.72 (1H, t, J
= 7.71 Hz, H-7), 8.08−8.22 (4H, m, H-5, H-8, H-2′ and H-6′). Anal.
(C27H32N2O4) C, H, N.
2-{[2-(4′-Propoxyphenyl)-4-quinolinyl]oxy}ethylamine (3y). To a
solution of 2-(2-{[2-(4′-propoxyphenyl)-4-quinolinyl]oxy}ethyl)-1H-
isoindole-1,3(2H)-dione (3w) (0.10 g, 0.22 mmol) in EtOH (10 mL)
was added 98% hydrazine hydrate solution (0.33 g, 0.66 mmol), and the
reaction was refluxed for 8 h with stirring. The reaction mixture was
cooled and filtered. The filtrate was evaporated under reduced pressure
to obtain a residue that was purified by flash chromatography (silica gel/
CHCl3/MeOH 99/1) to give compound 3y (0.01 g, 14% yield) as a
yellowish oil. 1H NMR (DMSO-d6): δ 1.03 (3H, t, J = 7.41 Hz,
OCH2CH2CH3), 1.68−1.80 (2H, m, OCH2CH2CH3), 3.50−3.65 (2H,
m, OCH2CH2NH2), 3.70−3.85 (2H, m, OCH2CH2NH2), 4.04 (2H, t, J
= 6.48 Hz, OCH2CH2CH3), 7.01 (1H, s, H-3), 7.10 (2H, d, J = 8.77 Hz,
H-3′and H-5′), 7.40−7.48 (3H, m, H-6 and NH2), 7.69 (1H, t, J = 7.69
Hz, H-7), 7.89 (1H, d, J = 8.45 Hz, H-8), 8.15 (2H, d, J = 8.84 Hz, H-
2′and H-6′), 8.28 (1H, d, J = 8.46 Hz, H-5). Anal. (C20H22N2O2) C, H,
N.
1-(2-{[2-(4′-Propoxyphenyl)quinolin-4-yl]oxy}ethyl)piperidin-4-
one (3z). To a solution of 4-[2-(1,4-dioxa-8-azaspiro[4.5]dec-8-
yl)ethoxy]-2-(4′-propoxyphenyl)quinolone 3x (0.07 g, 0.16 mmol) in
MeOH (2 mL) was added a HCl solution (9% in H2O, 2 mL), and the
reaction was refluxed for 6 h with stirring. The reaction mixture was
cooled and filtered. The solid obtained was then dissolved in CHCl3 and
dried over Na2SO4. The solvent was removed in vacuo to yield
compound 3z as a white solid (0.025g, 40% yield, mp 204.4−206.0 °C).
1H NMR (MeOD-d4): δ 1.10 (3H, t, J = 7.33 Hz, OCH2CH2CH3), 1.83-
4-[2-(1-Piperazinyl)ethoxy]-2-(4′-propoxyphenyl)quinoline (3q).
Obtained from piperazine as a brownish solid after purification by
flash chromatography (silica gel/CH2Cl2/MeOH 98/2) (30% yield, mp
1
121.8−123.1 °C). H NMR (CDCl3): δ.1.15 (3H, t, J = 7.34 Hz,
OCH2CH2CH3), 1.80−2.00 (2H, m, OCH2CH2CH3), 2.03 (1H, s,
NH), 2.75−2.85 (2H, m, OCH2CH2N), 3.07 (8H, t, J = 4.08 Hz,
piperazine CH2), 4.05 (2H, t, J = 6.62 Hz, OCH2CH2CH3), 4.46 (2H, t, J
= 5.63 Hz, OCH2CH2N), 7.07 (2H, d, J = 8.85 Hz, H-3′ and H-5′), 7.31
(1H, s, H-3), 7.50 (1H, t, J = 7.00 Hz, H-6), 7.74 (1H, t, J = 5.41 Hz, H-
7), 8.10 (3H, m, H-8, H-2′ and H-6′), 8.18 (1H, d, J = 8.70 Hz, H-5).
Anal. (C24H29N3O2) C, H, N.
4-[2-(4-Methyl-1-piperazinyl)ethoxy]-2-(4′-propoxyphenyl)-
quinoline (3r). Obtained from 1-methylpiperazine as white a solid after
crystallization from cyclohexane (28% yield, mp 108.7−110.1 °C). 1H
NMR (CDCl3): δ 1.11 (3H, t, J = 7.40 Hz, OCH2CH2CH3), 1.78−2.00
(2H, m, OCH2CH2CH3), 2.35 (3H, s, NCH3), 2.45−2.65 (4H, m,
piperazine CH2), 2.70−2.85 (4H, m, piperazine CH2), 3.07 (2H, t, J =
5.74 Hz, OCH2CH2N), 4.05 (2H, t, J = 6.59 Hz, OCH2CH2CH3), 4.47
(2H, t, J = 5.67 Hz, OCH2CH2N), 7.07 (2H, d, J = 8.86 Hz, H-3′and H-
5′), 7.19 (1H, s, H-3), 7.49 (1H, t, J = 6.93 Hz, H-6), 7.72 (1H, t, J = 6.95
Hz, H-7), 8.05−8.15 (3H, m, H-8, H-2′and H-6′), 8.20 (1H, d, J = 8.32
Hz, H-5). Anal. (C25H31N3O2) C, H, N.
4-[2-(4-Benzylpiperazin-1-yl)ethoxy]-2-(4′-propoxyphenyl)-
quinoline (3s). Obtained from 1-(phenylmethyl)piperazine as a yellow
solid after purification by flash chromatography (silica gel/PE/EtOAc
1
80/20) (43% yield, mp 102.5−103.3 °C). H NMR (CDCl3): δ 1.08
(3H, t, J = 7.26 Hz, OCH2CH2CH3), 1.81−1.92 (2H, m,
OCH2CH2CH3), 2.54−2.57 (4H, m, piperazine CH2), 2.72−2.74
(4H, m, piperazine CH2), 3.03 (2H, t, J = 5.85 Hz, OCH2CH2N), 3.55
(2H, s, CH2Ph), 4.02 (2H, t, J = 6.55 Hz, OCH2CH2CH3), 4.43 (2H, t, J
= 5.92 Hz, OCH2CH2N), 7.04 (2H, d, J = 8.83 Hz, H-3′ and H-5′), 7.15
(1H, s, H-3), 7.27−7.33 (5H, m, phenyl CH), 7.45 (1H, t, J = 8.80 Hz,
H-6), 7.68 (1H, t, J = 6.89 Hz, H-7), 8.05−8.17 (4H, m, H-5, H-8, H-2′
and H-6′). Anal. (C31H35N3O2) C, H, N.
N,N-Dibenzyl-2-{[2-(4′-propoxyphenyl)quinolin-4-yl]oxy}-
ethanamine (3t). Obtained from bis(phenylmethyl)amine as a white
solid after purification by flash chromatography (silica gel/PE/EtOAc
95/5) (30% yield, mp 76.6−78.4 °C). 1H NMR (CDCl3): δ 1.11 (3H, t,
J = 7.43 Hz, OCH2CH2CH3), 1.85−1.95 (2H, m, OCH2CH2CH3), 4.05
(2H, t, J = 6.59 Hz, OCH2CH2CH3), 4.40−4.56 (6H, m, OCH2CH2N
and CH2Ph), 4.80 (2H, t, J = 3.95 Hz, OCH2CH2N), 7.02−7.46 (14H,
m, H-3, H-6,H-3′, H-5′ and phenyl CH), 7.73 (1H, t, J = 7.63 Hz, H-7),
8.09−8.23 (4H, m, H-5, H-8, H-2′ and H-6′). Anal. (C34H34N2O2) C, H,
N.
Dibutyl(2-{[2-(4′-propoxyphenyl)quinolin-4-yl]oxy}ethyl)amine
(3u). Obtained from dibutylamine as a white solid after purification by
flash chromatography (silica gel/CH2Cl2/Acetone 97/3) (35% yield,
mp 64.0−65.7 °C). 1H NMR (CDCl3): δ 0.97 (6H, t, J = 7.17 Hz, 2 ×
NCH2CH2CH2CH3), 1.11 (3H, t, J = 7.31 Hz,OCH2CH2CH3,), 1.26−
2.02 (2H, m, OCH2CH2CH3), 2.08−2.14 (4H, m, piperidinone CH2N),
3.13−3.32 (4H, m, piperidinone CH2CO), 3.90−3.98 (2H, m,
OCH2CH2N), 4.27 (2H, t, J = 6.42 Hz, OCH2CH2 CH3), 5.00−5.07
(2H, m, OCH2CH2N), 7.26 (2H, d, J = 8.97 Hz, H-3′ and H-5′), 7.74
L
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(1H, s, H-3), 7.87 (1H, t, J = 7.98 Hz, H-6), 8.07−8.18 (3H, m, H-7, H-
2′ and H-6′), 8.24 (1H, d, J = 7.91 Hz, H-8), 8.58 (1H, d, J = 8.67 Hz, H-
5). Anal. (C25H28N2O3) C, H, N.
REFERENCES
■
(1) Spellberg, B.; Guidos, R.; Gilbert, D.; Bradley, J.; Boucher, H. W.;
Scheld, W. M.; Bartlett, J. G.; Edwards, J., Jr. The epidemic of antibiotic-
resistant infections: a call to action for the medical community from the
Infectious Diseases Society of America. Clin. Infect. Dis. 2008, 46, 155−
164.
(2) Boucher, H. W.; Talbot, G. H.; Bradley, J. S.; Edwards, J. E.; Gilbert,
D.; Rice, L. B.; Scheld, M.; Spellberg, B.; Bartlett, J. Bad bugs, no drugs:
no ESKAPE! An update from the Infectious Diseases Society of
America. Clin. Infect. Dis. 2009, 48, 1−12.
(3) Standing up to antimicrobial resistance. Nature. Rev. Microbiol.
2010, 8, 836.
(4) Taubes, G. The bacteria fight back. Science 2008, 356−361.
(5) Chambers, H. F. Community-associated MRSAresistance and
virulence converge. N. Engl. J. Med. 2005, 352, 1485−1487.
(6) Moran, G. J.; Krishnadasan, A.; Gorwitz, R. J.; Fosheim, G. E.;
McDougal, L. K.; Carey, R. B.; Talan, D. A. Methicillin-resistant S. aureus
infections among patients in the emergency department. N. Engl. J. Med.
2006, 355, 666−674.
(7) Lowy, F. D. Staphylococcus aureus infections. N. Engl. J. Med. 1998,
339, 520−532.
(8) Carbon, C. MRSA and MRSE: is there an answer? Clin. Microbiol.
Infect. 2000, 6 (Suppl 2), 17−22.
(9) Klein, E.; Smith, D. L.; Laxminarayan, R. Hospitalizations and
deaths caused by methicillin-resistant Staphylococcus aureus, United
States, 1999−2005. Emerg. Infect. Dis. 2007, 13, 1840−1846.
(10) Bancroft, E. A. Antimicrobial resistance: it’s not just for hospitals.
JAMA, J. Am. Med. Assoc. 2007, 298, 1803−1804.
(11) The Bacterial Challenge: Time to React; ECDC/EMEA Joint
Working Group, 2009; http://www.ecdc.europa.eu/en/publications/
Publications/0909_TER_The_Bacterial_Challenge_Time_to_React.
pdf (accessed June 7, 2013).
Bacterial Strains. The strains of S. aureus employed included ATCC
25923 (wild-type), SA-K2378, which overexpresses norA from a
multicopy plasmid and was produced by cloning norA and its promoter
into plasmid pCU1 and then introducing the construct into SA-K1902
(norA−).59 In addition, SA-1199B (overexpressing norA and also
possessing an A116E GrlA substitution) and its isogenic parent SA-1199
(norA wt) were also used.19,51 SA-K2885 and K2886 are norA-deleted
strains containing the empty expression vector pALC2073 and
pALC2073-mepA, respectively.60 Genes cloned into pALC2073 are
under control of a xyl/tetO promoter, which is inducible by 0.05 μg/mL
tetracycline.61 This concentration of tetracycline was included in all
experiments utilizing these strains.
EtBr Efflux. The loss of EtBr from S. aureus SA-1199B (norA
+/A116EGrlA) was determined fluorometrically as previously
described.50 With the exception of reserpine, which was replicated
numerous times, experiments were usually performed in duplicate, and
the results were expressed as mean total efflux over a 5 min time course.
The effect of various concentrations of reserpine and tested compounds
on the EtBr efflux of SA-1199B (norA+/A116E GrlA) was compared to
that in their absence, allowing the calculation of the percentage
reduction in efflux. Dose−response experiments were performed by
testing increasing concentrations of inhibitors, and the 50% inhibitory
concentration (IC50) was determined by inspection of the resultant plots
(see Supporting Information).
Microbiologic Procedures. MICs were determined by micro-
dilution techniques according to CLSI guidelines.62 The effect of
combining reserpine or various test compounds, with scalar dilutions of
freshly prepared solutions of each selected compound, on the MICs of
CPX was also determined. Checkerboard combination studies using
CPX and tested compounds were performed as described previously.63
(12) David, M. Z.; Daum, R. S. Community-associated methicillin-
resistant Staphylococcus aureus: epidemiology and clinical consequences
of an emerging epidemic. Clin. Microbiol. Rev. 2010, 23, 616−687.
(13) Centers for Disease Control and Prevention. Public Health
Dispatch: Vancomycin-Resistant Staphylococcus aureusPennsylvania.
Morb. Mortal. Wkly. Rep. 2002, 51, 902.
(14) Lynch, A. S. Efflux systems in bacterial pathogens: an opportunity
for therapeutic intervention? An industry view. Biochem. Pharmacol.
2006, 71, 949−956.
(15) Marquez, B. Bacterial efflux systems and efflux pumps inhibitors.
Biochimie 2005, 87, 1137−1147.
(16) Patel, D.; Kosmidis, C.; Seo, S. M.; Kaatz, G. W. Ethidium
bromide MIC screening for enhanced efflux pump gene expression or
efflux activity in Staphylococcus aureus. Antimicrob. Agents Chemother.
2010, 54, 5070−5073.
(17) Pao, S. S.; Paulsen, I. T.; Saier, M. H., Jr. Major facilitator
superfamily. Microbiol. Mol. Biol. Rev. 1998, 62, 1−34.
(18) Yoshida, H.; Bogaki, M.; Nakamura, S.; Ubukata, K.; Konno, M.
Nucleotide sequence and characterization of the Staphylococcus aureus
norA gene, which confers resistance to quinolones. J. Bacteriol. 1990,
172, 6942−6949.
(19) Kaatz, G. W.; Seo, S. M. Mechanisms of fluoroquinolone
resistance in genetically related strains of Staphylococcus aureus.
Antimicrob. Agents Chemother. 1997, 41, 2733−2737.
(20) Neyfakh, A. A. Mystery of multidrug transporters: the answer can
be simple. Mol. Microbiol. 2002, 44, 1123−1130.
ASSOCIATED CONTENT
■
S
* Supporting Information
Structure of the “inactive set” of EPIs used to build the
pharmacophore model, fitness values of the “active set” of
compounds perfectly matching the APRR hypothesis, EtBr efflux
inhibition dose−response experiment plots used to determine
the 50% inhibitory concentration (IC50), and elemental analysis
data for target compounds. This material is available free of
charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*Phone: +39 075 5855130. Fax: +39 075 5855115. E-mail:
stefano.sabatini@unipg.it.
Author Contributions
The manuscript was written through contributions of all authors.
All authors have given approval to the final version of the
manuscript.
Notes
The authors declare no competing financial interest.
(21) Mahamoud, A.; Chevalier, J.; Alibert-Franco, S.; Kern, W. V.;
Pages, J. M. Antibiotic efflux pumps in Gram-negative bacteria: the
inhibitor response strategy. J. Antimicrob. Chemother. 2007, 59, 1223−
1229.
ACKNOWLEDGMENTS
■
We thank Dr. C. Guercini for helpful discussions and R. Bianconi
and C. Moriconi for their technical support.
(22) Stermitz, F. R.; Lorenz, P.; Tawara, J. N.; Zenewicz, L. A.; Lewis,
K. Synergy in a medicinal plant: antimicrobial action of berberine
potentiated by 5′-methoxyhydnocarpin, a multidrug pump inhibitor.
Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 1433−1437.
(23) Morel, C.; Stermitz, F. R.; Tegos, G.; Lewis, K. Isoflavones as
potentiators of antibacterial activity. J. Agric. Food Chem. 2003, 51,
5677−5679.
ABBREVIATIONS USED
■
MDR, multidrug resistance; MRSA, methicillin-resistant Staph-
ylococcus aureus; MFS, major facilitator superfamily; EPI, efflux
pump inhibitor; EtBr, ethidium bromide; CPX, ciprofloxacin;
LBDD, ligand-based drug design
M
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(24) Belofsky, G.; Percivill, D.; Lewis, K.; Tegos, G. P.; Ekart, J.
Phenolic metabolites of Dalea versicolor that enhance antibiotic activity
against model pathogenic bacteria. J. Nat. Prod. 2004, 67, 481−484.
(25) Wright, G. D. Resisting resistance: new chemical strategies for
battling superbugs. Chem. Biol. 2000, 7, R127−R132.
(26) Lomovskaya, O.; Lee, A.; Hoshino, K.; Ishida, H.; Mistry, A.;
Warren, M. S.; Boyer, E.; Chamberland, S.; Lee, V. J. Use of a genetic
approach to evaluate the consequences of inhibition of efflux pumps in
Pseudomonas Aeruginosa. Antimicrob. Agents Chemother. 1999, 43, 1340−
1346.
(27) Kvist, M.; Hancock, V.; Klemm, P. Inactivation of efflux pumps
abolishes bacterial biofilm formation. Appl. Environ. Microbiol. 2008, 74,
7376−7382.
(28) Schmitz, F. J.; Fluit, A. C.; Luckefahr, M.; Engler, B.; Hofmann, B.;
Verhoef, J.; Heinz, H. P.; Hadding, U.; Jones, M. E. The effect of
reserpine, an inhibitor of multidrug efflux pumps, on the in vitro
activities of ciprofloxacin, sparfloxacin and moxifloxacin against clinical
isolates of Staphylococcus aureus. J. Antimicrob. Chemother. 1998, 42,
807−810.
(29) Aeschlimann, J. R.; Dresser, L. D.; Kaatz, G. W.; Rybak, M. J.
Effects of NorA inhibitors on in vitro antibacterial activities and
postantibiotic effects of levofloxacin, ciprofloxacin, and norfloxacin in
genetically related strains of Staphylococcus aureus. Antimicrob. Agents
Chemother. 1999, 43, 335−340.
(30) Vidaillac, C.; Guillon, J.; Arpin, C.; Forfar-Bares, I.; Ba, B. B.;
Grellet, J.; Moreau, S.; Caignard, D. H.; Jarry, C.; Quentin, C. Synthesis
of omeprazole analogues and evaluation of these as potential inhibitors
of the multidrug efflux pump NorA of Staphylococcus aureus. Antimicrob.
Agents Chemother. 2007, 51, 831−838.
(31) Munoz-Bellido, J. L.; Munoz-Criado, S.; Garcia-Rodriguez, J. A.
Antimicrobial activity of psychotropic drugs: selective serotonin
reuptake inhibitors. Int. J. Antimicrob. Agents 2000, 14, 177−180.
(32) Kaatz, G. W.; Moudgal, V. V.; Seo, S. M.; Hansen, J. B.;
Kristiansen, J. E. Phenylpiperidine selective serotonin reuptake
inhibitors interfere with multidrug efflux pump activity in Staphylococcus
aureus. Int. J. Antimicrob. Agents 2003, 22, 254−261.
(33) Kaatz, G. W.; Moudgal, V. V.; Seo, S. M.; Kristiansen, J. E.
Phenothiazines and thioxanthenes inhibit multidrug efflux pump activity
in Staphylococcus aureus. Antimicrob. Agents Chemother. 2003, 47, 719−
726.
(42) Sabatini, S.; Gosetto, F.; Serritella, S.; Manfroni, G.; Tabarrini, O.;
Iraci, N.; Brincat, J. P.; Carosati, E.; Villarini, M.; Kaatz, G. W.; Cecchetti,
V. Pyrazolo[4,3-c][1,2]benzothiazines 5,5-dioxide: a promising new
class of Staphylococcus aureus NorA efflux pump inhibitors. J. Med. Chem.
2012, 55, 3568−3572.
(43) Brincat, J. P.; Carosati, E.; Sabatini, S.; Manfroni, G.; Fravolini, A.;
Raygada, J. L.; Patel, D.; Kaatz, G. W.; Cruciani, G. Discovery of Novel
Inhibitors of the NorA Multidrug Transporter of Staphylococcus aureus. J.
Med. Chem. 2011, 54, 354−365.
(44) Brincat, J. P.; Broccatelli, F.; Sabatini, S.; Frosini, M.; Neri, A.;
Kaatz, G. W.; Cruciani, G.; Carosati, E. Ligand Promiscuity between the
Efflux Pumps Human P-Glycoprotein and S. aureus NorA. ACS Med.
Chem. Lett. 2012, 3, 248−251.
(45) Pieroni, M.; Sabatini, S.; Massari, S.; Kaatz, G. W.; Cecchetti, V.;
Tabarrini, O. Searching for innovative quinolone-like scaffolds:
synthesis and biological evaluation of 2,1-benzothiazine 2,2-dioxide
derivatives. Med. Chem. Commun. 2012, 3, 1092−1097.
́
(46) Handzlik, J.; Matys, A.; Kiec-Kononowicz, K. Recent Advances in
Multi-Drug Resistance (MDR) Efflux Pump Inhibitors of Gram-Positive
Bacteria S. aureus. Antibiotics 2013, 2, 28−45.
(47) Phase version 3.3; Schrodinger, LLC: New York, 2011.
̈
(48) Kaatz, G. W.; Moudgal, V. V.; Seo, S. M.; Hansen, J. B.;
Kristiansen, J. E. Phenylpiperidine selective serotonin reuptake
inhibitors interfere with multidrug efflux pump activity in Staphylococcus
aureus. Int. J. Antimicrob. Agents 2003, 22, 254−261.
(49) Michalet, S.; Cartier, G.; David, B.; Mariotte, A. M.; Dijoux-
Franca, M. G.; Kaatz, G. W.; Stavri, M.; Gibbons, S. N-
Caffeoylphenalkylamide derivatives as bacterial efflux pump inhibitors.
Bioorg. Med. Chem. Lett. 2007, 17, 1755−1758.
(50) Kaatz, G. W.; Seo, S. M.; O’Brien, L.; Wahiduzzaman, M.; Foster,
T. J. Evidence for the existence of a multidrug efflux transporter distinct
from NorA in Staphylococcus aureus. Antimicrob. Agents Chemother. 2000,
44, 1404−1406.
(51) Price, C. T.; Kaatz, G. W.; Gustafson, J. E. The multidrug efflux
pump NorA is not required for salicylate-induced reduction in drug
accumulation by Staphylococcus aureus. Int. J. Antimicrob. Agents 2002,
20, 206−213.
(52) QikProp, version 3.4; Schrodinger, LLC: New York, 2011.
̈
(53) Neyfakh, A. A. The Multidrug Efflux Transporter of Bacillus
subtilis Is a Structural and Functional Homolog of the Staphylococcus
NorA Protein. Antimicrob. Agents Chemother. 1992, 36, 484−485.
(34) Gibbons, S.; Oluwatuyi, M.; Kaatz, G. W. A novel inhibitor of
multidrug efflux pumps in Staphylococcus aureus. J. Antimicrob.
Chemother. 2003, 51, 13−17.
(54) MacroModel, version 9.9; Schrodinger, LLC: New York, 2011.
̈
(35) Gibbons, S. Anti-staphylococcal plant natural products. Nat. Prod.
Rep. 2004, 21, 263−277.
(55) Carpio, H.; Galeazzi, E.; Greenhouse, R.; Guzman
́
, A.; Velarde, E.;
Antonio, Y.; Franco, F.; Leon, A.; Perez, V.; Salas, R.; Valdes
́
́
, D.; Ackrell,
(36) Stavri, M.; Piddock, L. J.; Gibbons, S. Bacterial efflux pump
inhibitors from natural sources. J. Antimicrob. Chemother. 2007, 59,
1247−1260.
J.; Cho, D.; Gallegra, P.; Halpern, O.; Koehler, R.; Maddox, M. L.;
Muchowski, M. J.; Prince, A.; Tegg, D.; Thurber, T. C.; Van Horn, A. R.;
Wren, D. Synthesis of l,2-dihydro-3H-pyrrolo[1,2-a]pyrrole-1-carbox-
ylic acids and homologous pyridine and azepine analogues thereof. Can.
J. Chem. 1982, 60, 2295−2312.
(37) Guz, N. R.; Stermitz, F. R.; Johnson, J. B.; Beeson, T. D.; Willen,
S.; Hsiang, J.; Lewis, K. Flavonolignan and flavone inhibitors of a
Staphylococcus aureus multidrug resistance pump: structure−activity
relationships. J. Med. Chem. 2001, 44, 261−268.
(38) Samosorn, S.; Bremner, J. B.; Ball, A.; Lewis, K. Synthesis of
functionalized 2-aryl-5-nitro-1H-indoles and their activity as bacterial
NorA efflux pump inhibitors. Bioorg. Med. Chem. 2006, 14, 857−865.
(39) Sabatini, S.; Kaatz, G. W.; Rossolini, G. M.; Brandini, D.; Fravolini,
A. From phenothiazine to 3-phenyl-1,4-benzothiazine derivatives as
inhibitors of the Staphylococcus aureus NorA multidrug efflux pump. J.
Med. Chem. 2008, 51, 4321−4330.
(40) Pieroni, M.; Dimovska, M.; Brincat, J. P.; Sabatini, S.; Carosati, E.;
Massari, S.; Kaatz, G. W.; Fravolini, A. From 6-Aminoquinolone
Antibacterials to 6-Amino-7-thiopyranopyridinylquinolone Ethyl Esters
as Inhibitors of Staphylococcus aureus Multidrug Efflux Pumps. J. Med.
Chem. 2010, 53, 4466−4480.
(41) Sabatini, S.; Gosetto, F.; Manfroni, G.; Tabarrini, O.; Kaatz, G.
W.; Patel, D.; Cecchetti, V. Evolution from a Natural Flavones Nucleus
to Obtain 2-(4-Propoxyphenyl)quinoline Derivatives As Potent
Inhibitors of the S. aureus NorA Efflux Pump. J. Med. Chem. 2011, 54,
5722−5736.
(56) Maestro, version 9.2; Schrodinger, LLC: New York, 2011.
̈
(57) CombiGlide, version 2.7; Schrodinger, LLC: New York, 2011.
̈
(58) Spinks, D.; Armer, R. E.; Miller, D. J.; Rankovic, Z.; Spinks, G.;
Mestres, J.; Jaap, D. R. Preparation of 1-arylsulfonyl-3-substituted
indoles and indolines for the treatment of central nervous system
disorders. PCT Int. Appl. WO 2003068220 A1 20030821, 2003.
(59) Augustin, J.; Rosenstein, R.; Wieland, B.; Schneider, U.; Schnell,
N.; Engelke, G.; Entian, K. D.; Gotz, F. Genetic analysis of epidermin
biosynthetic genes and epidermin-negative mutants of Staphylococcus
epidermidis. Eur. J. Biochem. 1992, 204, 1149−1154.
(60) Kaatz, G. W.; McAleese, F.; Seo, S. M. Multidrug resistance in
Staphylococcus aureus due to overexpression of a novel multidrug and
toxin extrusion (MATE) transport protein. Antimicrob. Agents Chemo-
ther. 2005, 49, 1857−1864.
(61) Bateman, B. T.; Donegan, N. P.; Jarry, T. M.; Palma, M.; Cheung,
A. L. Evaluation of a tetracycline-inducible promoter in Staphylococcus
aureus in vitro and in vivo and its application in demonstrating the role of
sigB in microcolony formation. Infect. Immun. 2001, 69, 7851−7857.
N
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX

Journal of Medicinal Chemistry
Article
(62) Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria
That Grow Aerobically,7th ed.; Approved Standard M7-A7; Clinical and
Laboratory Standards Institute: Wayne PA, 2006.
(63) Eliopoulos, G. M.; Moellering, R. C. J. Antimicrobial
combinations. In Antibiotics in Laboratory Medicine; Lorian, V., Ed.;
Williams and Wilkins: Baltimore, MD. 1991; pp 432−492.
O
dx.doi.org/10.1021/jm400262a | J. Med. Chem. XXXX, XXX, XXX−XXX